The present invention refers to a method and a system for generating tag proposals for tagging of a content.
Tagging refers to assigning of significant keywords or meta-data, respectively, to contents, information, or data, respectively, (e.g., images, videos, product specifications, process or event specifications, etc.) available or provided in computing systems or apparatus. In the following, these keywords or meta-data, respectively, are referred to as “tags”.
The assigning of tags is usually performed based on subject matter and context of the corresponding content, to which the tags are assigned. Tags can be associated with regard to the whole content or with regard to parts of the content in order to provide an annotation of the content. In this way, efficient content management features such as searching and organization features, for example, are facilitated.
Accordingly, tags may be used to provide an automated mechanism, by which content management can be enhanced and user efforts can be minimized. Additionally, the performance of the apparatus or the system configured for managing of contents is improved in this way.
However, usually tags are standardized dependent upon several factors such as context or subject matter of contents, for example. Thus, tagging of contents or parts of a content that may have, for example, more than one context and/or more than one subject matter becomes rapidly more complicated. This, in turn, leads to a loss of efficiency of tagging and to a loss of performance of the respective content management. Accordingly, also the performance of an apparatus or a system configured for managing of contents by use of tagging decreases. Furthermore, a user typically has limited control of the context and therefore a limited control of the tagging of content items according to desires of the user.
Moreover, tags proposed by known tagging methodologies and to be associated to corresponding contents still lack quality and the required significance to specify the corresponding contents correctly. The known tagging methodologies can be affected or influenced by typographic and/or semantic errors. This leads to a lack of quality of the tags proposed by the known tagging methodologies. Sources of such errors are, for example, orthography (e.g., use of expressions in singular versus use of expressions in plural, upper and lower cases, etc.); use of separators for combinations of words; use of acronyms; use of synonyms (of different words with the same or similar meaning); use of different languages; use of several levels of abstraction of terms; use of homonyms (words with the same or similar spelling and pronunciation but with different meanings).
The lack of quality of tags proposed and assigned to contents is a further factor, leading to an insufficient management of contents and of modules, systems, and apparatus responsible for performing the management of contents.